as far as they selling out and doing it to make the most $$ they can easier.. you are right.. the games got simpler so they could basically use the same engine as I had said.. PCs are now stuck with the consoles scaled down engine.. with just a few small GFX enhancements for PC.. they have literally abandoned their original fans// PC gamers ... to make the bigger buck in sales volume to console players...
That Primal rating.Originally Posted by BridgeToClarity Go to original post
![]()
After you posted about it being a console issue ( I wouldn't know myself) I read up on it. There is a guy named Andrew Alvarez, a professional physics programmer for Gear Box. He made a very articulate post about "Why don't we see many physics-orientated games anymore?"Originally Posted by crunc01 Go to original post
I took it like it takes too much time, money, and can cause other problems in a game but in Battlefield we have the buildings getting blown to dust and things are able to be destroyed, on console as well. Despite BF5 being garbage imo. They do have great physics even on console.
Andrew Alvarez -
"Hi there, professional physics programmer for video games here to chime in. Basically, the reasons boil down to only a few main things all stemming from the inherent complexity involved in physics simulations. The complexity means that the simulation has the potential to be unpredictable and result in unforeseen consequences in development, performance, and player experience. For the latter, take your example of HL2 and go watch a speed-run of it. The amount of rigid body physics exploits used is staggering and they, without a doubt, break the game in so many unexpected ways. Generally speaking, developers try to avoid having those kinds of things in their games. Additionally, physics is hard. Ask anybody who has ever written a AAA quality physics engine from scratch and they'll tell you the same. There are so many things that could go wrong in so many different ways, not just in code but in content, and I don't think there is any physics engine that prevents all of these bad things from happening. Ask anybody who has tried to use a physics engine to accomplish a seemingly "simple" task and, chances are, they will be able to tell you ten different ways that it could be/is broken that they had to take into account. Usually, it's just easier to fake it when it comes to "physically-based" game mechanics. Physics can be unpredictable and players generally want things to happen how they expect them to.
So, this all leads to another point. The time investment. Physics related features in games are generally pretty time consuming. Even a character's ragdoll can take 20 or 30 hours of tuning to get to production quality. Additionally, each character that has a different rig will need a different ragdoll asset tuned for it. Add to that other "standard" features often expected in games such as cloth or other physically driven character assets (such as chains, hair, etc) and there ends up being potentially thousands of man hours of necessary tuning for these simple, "standard" features that usually don't even have much of an impact on gameplay. This means that we often get away with sub-optimally tuned physics. Case in point, look closely at the dynamic bits on characters in SFV. They actually jitter and clip quite a bit, but those things are generally considered acceptable artifacts. Jittering on a stack of boxes that the player builds and uses to climb to an upper-level or something isn't acceptable since it could cause the stack to fall over and block progress. So, now there is a bunch of extra time that goes into testing a bunch of things to makes sure the constraint solver iterates over these boxes enough to keep them stable, that there is no way any of these boxes could tunnel out of or otherwise get out of the play area and block progress, or any number of other issues that could occur. So, now we get back to just faking it. Don't make the boxes dynamic. Make them kinematic, do some basic collision tests to make sure they're not clipping into anything when they're put down/being held, and skip worrying about a million things being able to break since there are already a million other things broken.
Next thing is networking. This isn't applicable to all games but is still a major point to make. Networking is hard. Making sure physically simulated objects maintain parity across clients is also hard. And it can be a bit expensive. Look at Crackdown 3 and their destruction, for example. Absolutely amazing technology they've developed for cloud-computing the destruction for their multiplayer. That being said, it's not only computationally expensive but also has server overhead that results in extra costs in maintaining enough servers for all the multiplayer sessions since it's no longer 1 thread per session. So, this is a point not a lot of people who haven't been in the industry think about. The more lightweight a dedicated multiplayer server is, the more servers that can be run by a single CPU. If the dedicated server can manage to run on only 1 thread, then you can get N number of sessions running for the cost of one computational server where N is the number of threads that can be run. So, let's say the computer running the multiplayer server has 64 threads available to it. That single computer can run 64 multiplayer instances. A dedicated server that needs 2 threads to run, however, can only have 32 instances on that same computer. Now, to host the same number of sessions, you need twice as many computers. Now the publisher/developer/someone has to pay for that extra computer. So, when a core selling point of your game isn't "let's destroy a city with physics" then it make very little sense to dedicate a bunch of processing power to a feature like that since it will take away processing power from other features that are more relevant to the core of the game. It's the idea of, "just because we can, doesn't mean we should."
That's another big thing. "Just because we can, doesn't mean we should." Let's say we can blow up walls in our game. It's not a super difficult feature for physics to have (relatively speaking, of course). But now being able to blow up a wall has an impact on other gameplay systems. AI, for example, now need to be able to path through that blown up wall. Networking means that blown up wall needs to be replicated to all the other people in the session. For rendering, now we can't bake lighting for that wall because it has the potential to not be there anymore. Performance hits when that wall is destroyed because the number of dynamic objects goes up, the network traffic goes up, and the cost of rendering goes up. Is it really worth all the effort to blow up that wall? Does our game need you to be able to blow up that wall to deliver on one of the core pillars of the game? No? Then it's not worth it.
So, TL;DR is that, unless one of the core tenants of your game's design is "physically driven [blank]", it's just not worth the hassle on top of all the other physically driven stuff that is expected. Physics-based games like Fantastic Contraption, Angry Birds, and Goat Simulator still exist. However, when "physics-based" isn't a core mechanic, it's usually better to keep the physics stuff aesthetic and not too impactful to gameplay."
https://www.reddit.com/r/Games/comme...entated_games/
That all being said. The Far Cry games have some BAD physics compared to a lot of other games. The whole kill an animal and it starts sliding down a hill for no reason is really bad and has nothing to do with consoles. It doesn't happen in other games. Bodies disappearing after you kill them. I haven't seen that happen in other games that I play. I play on PS, Xbox 1 and a few PC games.
Originally Posted by Iamcuddles Go to original post
Sorry, lol
The first 10 hours of my play through I felt like it was a 9. The last 30 hours I wanted to throw my tv out the window.
Ubi love disappearing corpses and circus animal animations. Even their tactical shooter Ghost Recon Wildlands has corpses that disappear in 30 sec, or as little as 10 if you just look the other direction. It's claimed to be a "clean up" feature, supposedly to avoid making it very hard to remain undetected, but I can't help but feel it's a cheap way to keep FPS stable for those on mediorce rigs by limiting the active NPC AI scripts running.Originally Posted by BridgeToClarity Go to original post
Seeing bears roll down a hill at first seems funny, and my fave used to be boars that roll face first on their nose when you kill them charging you, but after too much of it the game reeks of novelty more than immersion.
Just when we thought EA was the worst. UBISOFT took the title back for FAR CRY 3. Each game after is a cash grab as well. Heck we get two boring game modes for map editor for FAR CRY 5. What a joke. Also no respawning vehicles. Waste of $60.
Don't expect people to buy FAR CRY 6 when you cant even fix your older games. Yet other games adding content all the time. Yet we get one patch other day of 9.28GB of bogus bug fixes that are still broken.
All UBISOFT games are so repetitive as well. Make something new. Also add more players for multiplayer. This is 2019 after all. 5vs5 for FAR CRY 5 yet FAR CRY 2 on last gen was 8vs8. Quit back peddling. If ya go skimpy, i'll go skimpy with my wallet.
Bad enough ya made New Dawn yet FAR CRY 5 is buggy and very limited like it was rushed and ditched after the little DLC's. Map editor could use some more content like new game modes and respawning vehicles. Also they didn't fix clipping errors or the redhorn8000B bug.
I'm not on PC but I agree with spending the time creating the single player aspect. I don't care for PVP and there are already tons of PVP games out there. Far Cry is pretty kick *** open world single player. I didn't play any of the PVP in FC5.Originally Posted by crunc01 Go to original post